The Source for Java Technology Collaboration
User: Password:
Register | Login help    



Syndicate this blog (XML)
Richard Monson-Haefel

Richard Monson-Haefel's Blog

«13 improvements for EJB | Main | JSR-241: Groovy – A New Standard Programming Language for the Java Platform »

The New York Times on Amazon.com Fraud

Posted by monsonhaefel on February 15, 2004 at 08:22 PM | Comments (10)

The New York Times recently published a story about authors anonymously posting bogus five star reviews on their own books. I was not mentioned in the article, but I was interviewed by the reporter, Ann Harman – she had read a blog I posted previously about fraud in the Amazon.com review process.Anyway, I think the reporter missed the main point. The issue is not anonymous reviews, it’s the use of fake identities to post bogus reviews.

There is, in my opinion, only one solution to Amazon.com's fraud-ridden book review system: Only customers who purchased the book from Amazon.com should be able to post a review on that book. Reviewers should be allowed anonymity, as long as Amazon.com keeps track of the reviewer's real identity. Amazon.com should not allow any Tom, Dick or Harry with a Yahoo! Or HotMail Email address to post reviews on any books they want – that's the system they have now and it has resulted in a review process that cannot be trusted. If you buy the book from Amazon.com, then you can review it.


Bookmark blog post: del.icio.us del.icio.us Digg Digg DZone DZone Furl Furl Reddit Reddit
Comments
Comments are listed in date ascending order (oldest first) | Post Comment

  • well said Or at the very least they should clearly mark all reviews from people that didn't buy the book through them as suspect. It is possible that someone purchased the book elsewhere after all :) Another option would be to have a kind of moderation feature that goes further than the "was this review helpful" they have now. If a set number of negatives is reached the review could be set to invisible until screened by an employee (or maybe a panel of volunteers from among people that purchased the item in the past).

    Posted by: jwenting on February 16, 2004 at 04:26 AM

  • Why do the reviews need to be anonymous? What are people afraid of when giving a bad review of a book when others know who they are? It works for eBay and makes it pretty clear when a seller or buyer is less than honest. When I am reading the reviews about books, I look at the content. If it just says: "This book sucks!", then I skip it and move to the next one. If the review has a pretty in depth discussion of the reasons for the reviews likes or dislikes, then I take the review to heart and move to the next. When I write reviews (very rare) I also post my reasoning behind my statements. Mind you, my feelings on a book may be completely different from someone eles, but at least someone knows that there is someone out there that feels a certain way about a book and they also know something about me and what I do for a living. I typically review technical manuals and programming language manuals. For an author to review his own work as if he/she was just a reader just to try to sell the book is wrong. For that author or Amazon to hire people to review titles is also wrong. Maybe instead of only allowing reviews from poeple who have bought the book from Amazon, the reviewers buying habits should be displayed. :) (I can see the privacy complaints just flooding in on that statement...) What that will show is the reviewer has bought other books from the same subject which would give me a better feeling about their review. Ex: If a review for a Java book was really bad and I saw that the reviewer only bought spy novels, I would ignore the review and move to the next. However, if they had purchased 10 other books on Java programming, I would know that they have a strong interest in the subject. Plus I could see if they had reviewed the other books to see if they had one that they liked instead. Typically, when I am looking to buy a book, I look at reviews in Blogs and also go to the newsgroups and see if anyone is talking about it out there. Much better than relying on Amazon's sucky review system. Sorry for the size of the post. I have a huge grudge with Amazon and specifically their cheesy review system.

    Posted by: vesperdem on February 16, 2004 at 09:08 AM

  • Identity of reviewers I suppose you could allow people to only review books they've bought through Amazon but that seems a bit extreme. I occasionally like to review a book I bought elsewhere. A half-way solution would be to only allow people who have bought books (or other items) through Amazon before. This way you know that they are a real person and are not impersonating anyone because you've sent them stuff that they've paid for before. On the other hand, especially in the IT field, there might be some consumer education to do. For instance, I always make a point of reading good and bad reviews on a book because I feel that the 1 star reviews usually tell you as much as the 5 star ones. I also disregard reviewers who don't qualify their argument, i.e., people who say it's great or it's awful without saying why.

    Posted by: brunogirin on February 16, 2004 at 09:27 AM

  • Reviewer type icon & use of poor reviews I'd have Amazon include one of two icons next to reviewers "names". One for people who have bought the book from Amazon (to satisfy Amazon's commercial interests). A second to indicate people had bought books of a related nature at Amazon (easy enough to implement; I can see a fun algorithm to do this in my head right now!). And nothing for those who fit neither category. The only catch is perhaps Amazon has done a survey and discovered, shock-horror, that most reviews are from people who did *not* buy the bok at Amazon! It'd look bad press to advertise this... I'd be for non-anonymous reviews, as long as the email address isn't directly on the page in a way that page scrapers can get at. We all get enough spam as it is... I agree that it pays to only use the reviews with detailed statements about the book in question, but with a twist. I look forreviews revealing the nature, or intended purpose, of the book and take the stars with a pinch of salt. I particularly look out for poorly rated reviews where the reviewer indicates (usually indirectly) that they thought the book was of one kind but is in fact another. For example, they expected a scientific text, but got a piece of lightweight popular science. Or expected a "how to" textbook, but got a reference. (It annoys me that people rate the book badly for this; its actually the marketing or their own judgement that is at fault, but all the same its useful to me!)

    Posted by: tramper on February 16, 2004 at 11:57 AM

  • Reviewer type icon & use of poor reviews I agree with the no e-mail address visible. There is no need. Just click on the name to see their "profile". Then you have control over what people see. I know I don't want my e-mail address in anyone's hands other than those I let know. Know, that said, with all the "problems" that Amazon has had with "security" of data. Maybe this is not such a good idea... :)

    Posted by: vesperdem on February 17, 2004 at 11:04 AM

  • Legitimate Amazon customers should be allowed to review Let me start by saying that I'm a long time fan of Richard Monson-Haefel's technology writing. However, I disagree with Richard's criteria for Amazon reviewers: "Only customers who purchased the book from Amazon.com should be able to post a review on that book." Two reasons. One. Once people buy a book they are already biased in its favor. Folks rarely wish to admit that they made a poor purchase. Two. Amazon can verify reviewers' identity as long as they've ever purchased a book from Amazon. So, the criteria should be that reviewers must be legitimate Amazon customers (with valid credit cards on file, which help to establish identity) even though they need not have purchased the book they're reviewing. What say? -- Puneet Singh Lamba

    Posted by: plamba on March 19, 2004 at 02:10 PM




This work is licensed under a Creative Commons License.
Creative Commons License

Powered by
Movable Type 3.01D
 Feed java.net RSS Feeds
Feedback  | FAQ  | Press | Terms of Use
Privacy  |  Trademarks | Site Map
 
Your use of this web site or any of its content or software indicates your agreement to be bound by these Terms of Participation.
 
Copyright © 1995-2008 Sun Microsystems, Inc.
Oreilly LogoCollabnet Logo
Powered by Sun Microsystems, Inc.,
O'Reilly and CollabNet